Revista de Economia Política 30 (3), 2010
463
PROCEDURAL RATIONALITY
These simplifications in the mechanisms of choice advanced by Simon — and
that, as we have seen, appeared in the mid‑fifties (Simon, 1955, 1956) — are ele‑
ments that bore already more importance to the decision making process and that
would, later, constitute the central components of the concept of procedural ratio‑
nality. Initially, the question of computation appears somewhat muted under the
idea of computational “capacity”, but it is present, as was pointed above. I say that
the idea of “capacity” hinders the complete development of computational issues,
because computation is something that has an important qualitative dimension,
and fundamentally procedural: the process is embodied in the program, which
describes the way computing is to be done. These computational concerns, there‑
fore, appear in Simon’s work simultaneously to the concept of satisficing, initially
labeled “satisfactory pay‑offs”. Satisficing is essentially the hypothesis that allows,
and practically induces, to the conception of diverse decision procedures. With it,
the decision maker does not have to take into account all possible behavior alterna‑
tives and, in addition, does not need to worry about ascertaining whether the al‑
ternatives he or she is considering are, in fact, all the possible ones. Alternatives
can be sequentially found out, by search processes, search being interrupted when
a satisfactory alternative is found. Satisficing is, hence, the theoretical step that
allows Simon to abandon the idea of rationality as a tautological reasoning over
given premises, which permits rationality to operate in an open, not predetermined,
space. On the other hand, satisficing forces him to inquire into the process by which
such premises are built by the agent. The point I wish to emphasize is that, in the
mid‑fifties, although it is not yet the idea of procedure that organizes Simon’s ef‑
forts, the need to theorize about the decision procedure is already implied in his
theoretical propositions. Moreover, and more importantly, starting from the critic
of the boundaries to global rationality, every attempt at positive construction educ‑
es the procedural dimension of decision making. However, bounded rationality is
always only the starting point and maintains its character of a construction in
negative: “in conditions of bounded rationality” the agents resort to other expedi‑
ents, different from those prescribed by global rationality, in order to exercise their
intention of rationality. The specification of such expedients, of other types of ra‑
tional behavior, is the reaction to a bounded rationality condition, but it is not
bounded rationality itself.
The fact that these two elements — computation and satisficing — appear in
Simon’s work in 1955 is not casual. In 1952, he became a consultant to RAND
Corporation, initially involved in simulations of an air‑defense early warning sta‑
tion, and then, from 1955 on, connected with the Computer Science Department.
RAND was the paradigmatic military think tank in the post‑Second World War
period. It was also the world’s largest computational structure for scientific ends
at the time. Simon’s entrance in RAND marks an intellectual inflection of his.
Among the aspects of this change that interests us here is his distancing away from
economics toward the areas of psychology and computer science, a move that
Revista de Economia Política 30 (3), 2010
464
would only, and partly, be reverted in the 1970s — more specifically he placed
himself in the nascent disciplines of cognitive psychology, cognitive science, artifi‑
cial intelligence, operations research, and computer science, all of them tightly
connected with the computer. His research program became essentially aimed at
discovering the symbolic processes that people use in thinking, and was based on
the exploration of an analogy between the computer and the human mind. The
main method used was the combination of the tape‑recording of the problem solv‑
ing activity of subjects in the laboratory — producing “thinking‑aloud protocols”
— and of the simulation of computer programs that tried to emulate the activity
registered in the laboratory. This meant that programs were taken to be theories:
the program capable of simulating the human behavior recorded in the laboratory
is, in itself, an explanation to that behavior. The attempt at programming (theoriz‑
ing) the solution processes of relatively complex problems in computers with very
limited memory and processing capacity led to the satisficing hypothesis, maximi‑
zation would be impracticable without drastic simplification of the model. In oth‑
er words, if, on the one hand, the mind‑computer analogy suggests a very concrete
image of what are the agents’ cognitive limits, on the other hand, programming
always demands specification: what information the agent possesses, what criteria
and procedures he or she uses to make decisions. Without such specifications, the
programming cannot even begin.
It is based on his work at RAND and his contact with computers, then, that
Simon starts to advance in a more positive manner other concepts of rationality,
which diverged from global rationality. These would later (1976b) be grouped
under the term “procedural rationality”, in an attempt at reinforcing the impor‑
tance of the decision making process to the theory. Still concerning this matter, it
is important to point that the very same basic theoretical elements that emerged in
the 1950s as “simplifications” of the global rationality model form the core of the
“procedures” in the 1970s, especially satisficing. Moreover, if the problems associ‑
ated with computation were already in the fifties the main source of positive ad‑
vances in the definition of rationality, they came to be central in the theory. More
detailed comment upon these two issues is due.
Simon himself, by the late seventies, considered two concepts — already clear‑
ly present in his interventions in the economic science field in the fifties, and which
we had the opportunity to discuss above — as the central elements to a more
“positive” characterization of the mechanisms of decision: satisficing and search.
In his own words:
In Administrative Behavior, bounded rationality is largely char‑
acterized as a residual category — rationality is bounded when it falls
short of omniscience. And the failures of omniscience are largely failures
of knowing all the alternatives, uncertainty about relevant exogenous
events, and inability to calculate consequences. There was needed a more
positive and formal characterization of the mechanisms of choice under